A high-quality question–answer dataset of 100 records focused on the Common
Vulnerability Scoring System (CVSS) Version 4.0. It is built to train and evaluate AI
systems that explain CVSS scoring methodology, interpret vector strings, justify scoring
decisions, evaluate vulnerability severity, and help analysts produce consistent,
defensible vulnerability assessments — LLM fine-tuning, retrieval-augmented generation
(RAG), vulnerability-management assistants… See the full description on the dataset page:
https://huggingface.co/datasets/ismailtasdelen/cvss-v4-qa.
